The Student Assistance Program (SAP) is the most popular early intervention for adolescent alcohol abuse, and preliminary data suggest that SAPs can be effective for reducing alcohol abuse in some students. However, conclusive statements about the effectiveness of SAPs cannot be made because of several methodological limitations of SAP studies performed to date. These studies have not: a) included a comparison group; b) engaged in systematic, repeated follow-up of students who take part in SAPs; or, c) evaluated students on standardized alcohol and/or other drug use measures. Furthermore, despite consistently finding that a sizable number of teens do not change their substance use in response to. SAPs, studies have not examined who does or does not benefit from SAP participation or what the "active ingredients" of SAPs are. The current study proposes to compare the effectiveness of a SAP group counseling (GC) intervention with assessment/referral-only (AR) for reducing alcohol abuse and related problems among adolescents. We will also assess the impact of depressive symptoms on substance use outcomes given evidence that depressed teens may demonstrate less response than nondepressed teens. Finally, we will examine the impact of other potential predictor variables (i.e., alcohol expectancies, problem behaviors, social support) on substance use outcomes. Major hypotheses are: a) GC will be more effective in achieving and maintaining reductions in alcohol and other drug use among alcohol abusing adolescents than AR; b) GC will be more effective in improving academic conduct among alcohol abusing adolescents than AR; and, c) alcohol abusing adolescents with more depressive symptoms will have poorer substance use outcomes than adolescents with fewer depressive symptoms. The results of this study will contribute significantly to the knowledge base concerning the effectiveness of school-based early interventions for adolescent alcohol abuse. Data will be obtained concerning a) the overall effectiveness of GC compared to AR and b) the "active ingredients" of GC intervention. Furthermore, the results will provide valuable new information concerning the predictors of substance use outcomes of alcohol abusing adolescents.